The receptor tyrosine kinase ERBB4 has unique activities in the breast. For example, the other three ERBB-family members, namely EGFR, ERBB2/HER2/Neu, and ERBB3, directly contribute to aggressive breast cancer. Despite providing an essential function during lactation, ERBB4 expression is selectively extinguished during the progression of breast cancer. Significantly, ERBB4 functions as a pro-apoptotic BH3-only protein, of the BCL-2 family, specifically inducing apoptosis of malignant but not normal breast epithelium; thus providing a mechanistic explanation for the selective loss of ERBB4 expression in breast carcinomas. In normal breast epithelium, ERBB4 is a nuclear protein and activates milk-gene expression by functioning as a nuclear chaperone for the transcription factor, STAT5A. These divergent cellular responses to ERBB4 require proteolytic processing and subsequent release from the cell membrane the ERBB4 endodomain. Interestingly, the ERBB4 endodomain harbors the nuclear localization signal essential for stimulation of milk-gene expression and the pro-apoptotic BH3 domain. These observations support the hypothesis that novel signaling pathways mediated by the membrane- released ERBB4 endodomain regulate divergent ERBB4 activities in the normal and neoplastic breast. The following specific aims were designed to test this hypothesis: Aim 1) Test the hypothesis that ERBB4 endodomain signaling directly contributes to breast development; Aim 2) Test the hypothesis that novel STAT5A phosphorylation events modulate ERBB4 endodomain-stimulated gene expression; Aim 3) Test the hypothesis that BCL-2-farnily members directly regulate ERBB4-induced apoptosis. The experiments described in this proposal are designed to provide a thorough analysis of nuclear and cytosolic ERBB4 endodomain signaling in the normal and neoplastic breast, respectively. Molecular mechanisms ofERBB4 fimction in the normal breast integrated with mechanistic studies of ERBB4 cell killing of malignant cells will provide the necessary foundation to develop ERBB4 signaling as a highly specific therapeutic approach for the treatment of breast tumors while sparing normal breast cells.